Inspiration:
Pharmaceutical companies today make critical decisions using delayed reports, surveys, and manual market analysis. This often causes them to react too late to changing doctor preferences, patient concerns, and competitor launches — leading to lost opportunities and revenue. We were inspired to build PulseRx to create an always-on AI system that continuously listens to the healthcare ecosystem, detects market shifts early, and recommends timely actions, enabling pharma teams to move from reactive decision-making to proactive intelligence.
What it does?
PulseRx continuously monitors healthcare news, medical forums, research publications, and pharma updates in real time. It uses AI to analyze sentiment, detect emerging trends, identify competitor threats, and predict market risks. The system then sends smart alerts and actionable recommendations to pharma teams, helping them respond faster and make better business decisions.
How we built it
We built PulseRx using a full-stack AI architecture. The backend ingests real-time data from healthcare news, research sources, and online discussions. This data is processed using NLP models for sentiment analysis and trend detection, then stored in a vector database for semantic search. A machine learning layer predicts market risks and opportunities. The frontend dashboard displays insights and triggers smart alerts with recommended actions. We used modern web frameworks and LLM-based agents to enable real-time intelligence and decision support
Challenges we ran into
We faced challenges in collecting clean and relevant healthcare data from multiple sources and handling noisy, unstructured text. Building accurate sentiment and trend detection with limited time and data was difficult. Designing meaningful predictions instead of just summaries was another challenge. We also worked on optimizing real-time processing and integrating multiple AI components smoothly within the hackathon timeline.
Accomplishments that we're proud of
We successfully built an always-on AI agent that converts raw healthcare data into real-time market insights and actionable recommendations. We implemented end-to-end data ingestion, sentiment analysis, trend detection, and predictive alerts within the hackathon timeframe. Most importantly, we delivered a working prototype that demonstrates real business value beyond dashboards or chatbots.
What we learned
We learned how to build real-time AI systems that process unstructured data at scale and convert insights into actionable business intelligence. We gained hands-on experience with NLP pipelines, AI agents, and predictive modeling. We also learned the importance of designing solutions around real business problems, not just technical features, and working effectively as a team under tight hackathon deadlines.
What's next for PulseRx
We plan to expand PulseRx by integrating more verified medical data sources, improving prediction accuracy using advanced time-series and graph models, and adding multilingual support for global markets. We aim to deploy real-time integrations with pharma CRM systems and build enterprise-grade compliance features, making PulseRx production-ready for large pharmaceutical organizations.
Log in or sign up for Devpost to join the conversation.